It is widely believed that there are educational benefits to making computer games, but there is no systematic review of research on this topic. This article describes a meta-synthesis of research on children designing and programming computer games that investigates the extent to which there is evidence of benefits for computer science learning and motivation. Over 400 articles were identified, and 68 articles met the inclusion criteria. A systematic analysis and synthesis across studies showed some evidence that computer game design and programming can lead to changes in programming knowledge, problem solving, and computer science attitudes and confidence. However, most of the evidence described engagement in computing-related practices and did not measure learning. The findings were mostly positive, although several studies noted more negative attitudes toward programming after making games. The results were similar across different pedagogical approaches, although social interaction may provide unique opportunities for computer science learning. The synthesis resulted in a list of design elements for studying computer game design and programming activities; these can be used to increase the availability of evidence about learning. The article concludes with the identification of gaps in the research and suggestions for additional research.